共查询到20条相似文献,搜索用时 78 毫秒
1.
《组合机床与自动化加工技术》2019,(9)
针对超低速滚动轴承故障诊断困难问题,提出一种自适应噪声的完备集合经验模态分解(CEEMDAN)与深度信念网络(DBN)相结合的超低速滚动轴承故障声发射(AE)诊断方法。通过EEMD和CEEMDAN方法分别对轴承AE信号进行分解,结果表明,CEEMDAN具有较好的分解完备性和抗模态混叠性;将EEMD能量熵和CEEMDAN能量熵分别作为模式识别分类器的特征向量进行故障诊断,后者的识别准确率较高;通过与SVM、BP神经网络方法对比,DBN方法的模式识别效果更好,且表现出较好的稳定性。因此,文章所提方法能够有效的应用于超低速滚动轴承的故障诊断。 相似文献
2.
《组合机床与自动化加工技术》2017,(3)
提出一种基于自适应噪声的完备经验模态分解(CEEMDAN)与IMF样本熵结合的滚动轴承故障特征提取方法。利用CEEMDAN算法对振动信号进行了自适应分解,将非稳定的振动信号分解成了若干个固有模态函数(IMF)分量。计算了包含主要故障特征信息的IMF分量样本熵,实现了故障特征量化。在此基础上利用SVM在少量数据样本的情况下具有较强的学习和分类能力,通过样本数据学习与待测样本的模式识别实现滚动轴承智能诊断。通过仿真与实验数据分析,证明该方法能够改善信号特征提取的效果,对故障类型的判断表现出较高的识别率。 相似文献
3.
《组合机床与自动化加工技术》2018,(11)
针对滚动轴承振动信号的非平稳特性,实际工况下难以采集大量的样本信号分析故障状态,提出基于自适应噪声的完备经验模态分解(CEEMDAN)与多尺度排列熵(MPE)相融合的故障识别方法。首先,对振动信号进行小波阈值去噪,利用CEEMDAN算法对去噪后的非平稳振动信号自适应分解,对分解后的若干个固有模式分量(IMF)计算互相关系数;然后,重构信号,计算其MPE并组成故障特征向量;最后,把特征向量输入到支持向量机(SVM)中,以识别滚动轴承的故障类型。通过对仿真信号以及实际实验数据的对比验证分析,有效证明了该方法的识别准确率比基于EMDMPE的故障识别方法提高5%,结果表明:基于CEEMDAN-MPE的滚动轴承SVM故障识别方法可以更准确地提取轴承的特征,并识别轴承的故障状态,有更强的实用性和有效性。 相似文献
4.
由于往复压缩机的振动信号具有非线性非平稳性的特点,为进一步提高故障识别率,提出对自适应噪声完备集合经验模态分解(CEEMDAN)进行改进并与复合层次散布熵相结合的往复压缩机气阀故障诊断方法。利用正交性为指标选择最佳模态函数,有效提高了CEEMDAN对非平稳性信号的分解精度,减少噪声残差;采用峭度作为评价指标对分解后的IMF分量进行筛选并重构信号,求解重构信号的复合层次散布熵,提取故障特征向量;利用支持向量机进行分类识别。试验结果验证了该方法的有效性和优越性。 相似文献
5.
6.
《组合机床与自动化加工技术》2019,(8)
针对齿轮箱轴承特征难以提取的问题,提出一种基于改进希尔伯特-黄变换(HHT)和形态学分形维数的故障特征提取方法。首先采用自适应白噪声总体经验模态分解(CEEMDAN)方法将轴承振动信号分解为若干个固有模态函数(IMF),然后分别计算各IMF分量的相关系数和峭度值以滤除对信号特征不敏感的分量,最后计算包含敏感故障特征分量所组成的重构信号的形态学分形维数,以此作为特征参数对轴承的工作状态进行识别。通过对实测轴承信号的分析,结果表明,文章所提方法可有效识别轴承的工作状态和故障类型。 相似文献
7.
滚动轴承早期故障信号易受噪声干扰,故障冲击成分难以提取,故障识别困难。为从多角度提取故障轴承振动信号特征参数,利用变分模态分解(VMD)将振动信号分解为若干本征模态分量(IMFs),基于包络熵、相关系数、峭度筛选IMF分量。提取所选IMF的时域和频域特征、信号VMD能量熵及各IMF能量比组成特征向量,从时域、频域和能量角度反映故障信息。使用麻雀搜索算法(SSA)优化SVM参数,确定最优参数,克服参数选择难题。将样本特征向量输入SSA-SVM中进行故障分类,轴承故障实验数据表明:该方法故障识别平均准确率在98.71%以上;与单一域特征相比,该方法对故障类型和损伤程度识别效果更佳。 相似文献
8.
针对滚动轴承故障信号的非线性特性及不同故障类型信号具有不同形态特征的特点,提出一种基于改进变分模态分解(VMD)形态谱和模糊C均值聚类(FCM)算法相结合的故障诊断方法。采用VMD方法对滚动轴承振动信号进行分解,针对分解过程中关键参数的选取,提出相关参数选择方法,并计算各固有模态函数(IMF)的能量波动系数,以获得对信号特征信息敏感的模态分量进行重构。计算重构信号的形态谱以反映信号的形态特征。通过FCM算法实现滚动轴承工作状态和故障类型的诊断。运用该方法对实测滚动轴承振动信号进行分析,并将所提方法同基于原始振动信号、经验模态分解、总体经验模态分解形态谱的故障特征提取方法进行对比。结果表明:所提方法能够更加有效提取滚动轴承信号的故障特征,实现故障类型的准确诊断。 相似文献
9.
针对强噪声下微小故障信号容易被噪声淹没的问题,提出基于最大二阶循环平稳盲解卷积(CYCBD)和自适应噪声完全集合经验模态分解(CEEMDAN)的轴承微小故障诊断方法。根据故障频率公式求出振动信号的故障频率,并根据故障频率设置对应的循环频率集,用CYCBD对原信号进行滤波,使信号中的周期冲击成分更加突出,从而达到提高信噪比的目的;对处理后的信号进行CEEMDAN,得到一系列模态分量,再求各模态分量的峭度值,从中选取峭度值高的即含有较多故障特征的若干分量进行重构;对重构后的信号求其Hilbert包络谱,从中提取故障频率。采用仿真信号与西储大学轴承数据集进行仿真与实验研究,验证所提方法的有效性。 相似文献
10.
针对齿轮箱轴承信号非平稳性及其故障特征难以提取的问题,提出一种自适应白噪声平均总体经验模态分解(CEEMDAN)能量熵和马氏距离相结合的故障诊断方法。首先采用CEEMDAN方法对非平稳的轴承故障信号进行分解,获得若干阶表征信号特性的固有模态函数(IMF)分量;然后计算各IMF分量的自相关函数和相关系数,以滤除信号内的噪声干扰和对故障特征不敏感的IMF分量;最后计算各敏感故障特征分量的能量熵,将其作为特征参数形成状态特征向量,并使用马氏距离判别方法对轴承的工作状态和故障类型进行诊断。通过对实测不同工况以及不同故障程度的齿轮箱轴承信号的分析,证明了所提方法的有效性。 相似文献
11.
一种适用于嵌入式数控系统的任务调度方法 总被引:1,自引:1,他引:0
对数控系统的多任务特殊性进行分析,并结合嵌入式运算资源的有限性,提出一种混合实时任务层次调度算法,在保证实时性的基础上,充分提高CPU的利用率,并给出该调度方法的层次调度模型和算法。 相似文献
12.
Application of a two-level optimization process to conceptual structural design of a machine tool 总被引:1,自引:0,他引:1
Bi-Chu Wu Gin-Shu Young Te-Yen Huang 《International Journal of Machine Tools and Manufacture》2000,40(6):783-794
This paper proposes a modified two-level optimization approach for the concept design of a machine tool. The lower level of optimization is applied to each structural part of the tool and the upper level to the machine tool as an integrated system. When frequent modification of design specifications is required and re-use of previous design experience is preferred, the proposed approach is efficient because of the approximation functions used. Examples presented in the paper show that the principal dimensions of all the structural parts can be determined, minimizing the weight of the machine while maintaining sufficient stiffness. 相似文献
13.
本文提出建立以组织级、过程控制级和执行组组成的FMS递阶智能控制系统来保证FMS运行的优化。建立了一个基于知识的调度专家系统和用于动态调度的人工智能启发式算法和BP神经网络为主要结构的FMS智能控制系统。控制系统采用状态实现多智能模块之间的协调和驱动。。 相似文献
14.
15.
This paper introduces HART-S, a new modular neural network (NN) that can incrementally learn stable hierarchical clusterings of arbitrary sequences of input patterns by self-organization. The network is a cascade of adaptive resonance theory (ART) modules, in which each module learns to cluster the differences between the input pattern and the selected category prototype at the previous module. Input patterns are first classified into a few broad categories, and successive ART modules find increasingly specific categories until a threshold is reached, the level of which can be controlled by a global parameter called 'resolution'. The network thus essentially implements a divisive (or splitting) hierarchical clustering algorithm: hence the name HART-S (for 'hierarchical ART with splitting'). HART-S is also compared and contrasted with HART-J (for 'hierarchical ART with joining'), another variant that was proposed earlier by the first author. The network dynamics are specified and some useful properties of both networks are given and then proven. Experiments were carried out on benchmark data sets to demonstrate the representational and learning capabilities of both networks and to compare the developed clusterings with those of two classical methods and a conceptual clustering algorithm. A brief survey of related NN models is also provided. 相似文献
16.
17.
18.
Christian R. Huyck 《连接科学》2007,19(1):1-24
Highly recurrent neural networks can learn reverberating circuits called Cell Assemblies (CAs). These networks can be used to categorize input, and this paper explores the ability of CAs to learn hierarchical categories. A simulator, based on spiking fatiguing leaky integrators, is presented with instances of base categories. Learning is done using a compensatory Hebbian learning rule. The model takes advantage of overlapping CAs where neurons may participate in more than one CA. Using the unsupervised compensatory learning rule, the networks learn a hierarchy of categories that correctly categorize 97% of the basic level presentations of the input in our test. It categorizes 100% of the super-categories correctly. A larger hierarchy is learned that correctly categorizes 100% of base categories, and 89% of super-categories. It is also shown how novel subcategories gain default information from their super-category. These simulations show that networks containing CAs can be used to learn hierarchical categories. The network then can successfully categorize novel inputs. 相似文献
19.
20.
A turtle carapace bio-inspired Ti matrix hybrid composite was successfully fabricated in this work.This composite incorporates two parts: the Ti–Al intermetallic multilayered composite and continuous Si C fibers-reinforced Ti matrix composite.In the Ti–Al intermetallic multilayered composite part,a series of Ti–Al intermetallics compounds,including Ti_3Al,Ti Al,Ti Al_2 and Ti Al_3,were formed between the Ti layers.In the continuous Si C fibers-reinforced Ti matrix composite part,Si C fibers and Ti matrix were found to be bonded well through weak interface reaction.Flexural strength of this material reached 1.21 ± 0.16 GPa,measured by three-point bending test.The deformation features suggest that the hierarchical structure combining ductile Ti layers/matrix with brittle high-strength Ti–Al intermetallics layers/Si C fibers can effectively enhance the mechanical properties of the bio-inspired hybrid composite. 相似文献